Hebbia: Meet the 25 year-old who just raised at a $700M valuation

(4 minute read)

• This AI startup is re-inventing the way we work

• Over the last 18 months, they’ve grown revenue 15x, and quintupled their headcount

• They just raised a $130M Series B at a $700M valuation led by a16z

So what’s the startup and who is the founder behind it? Here’s the story of Hebbia AI📈:

Hebbia was founded in 2020 by George Sivulka with the vision to completely reimagine the way we work.👨‍💻

Hebbia is dubbed as the “AI platform for knowledge work”. It uses the power of LLMs to automate workflows by breaking them down into understandable actions.

It’s essentially a productivity tool that has a suite of different LLM tools where ANYONE with no technical training whatsoever, can wrangle any technical documents to get any sort of output.

Users can use Hebbia to extract, structure, and analyze millions of documents AT SCALE using their gen-AI agents🤖.

💰Their product is already being used by 33%+ of the top 50 largest asset managers for due diligence, asset pricing and other research.

So how did George start out? 👶

As a teen, George had been dubbed a bit of a wonderkid, having worked for NASA 🚀 and finishing his Math Bachelor’s at Stanford in 2.5 years.

He was then a funded PhD student at Stanford studying math and was one of the only math students that actually wanted to teach for a living.

At Stanford, George was thrown into the infamous “Math 51” class 🤓, and got to see students there after they’d left and started their careers.

99% of anyone who took the class either ended up in FAANG or worked as a knowledge worker for banks like Morgan Stanley and Goldman🏦.

George realized that the most brilliant students would do this and wondered why they’d put themselves in such pain working long hours.

He knew this was a good place to build a company around but forgot about it...

George then started his career as a computational neuroscientist at Stanford and was constantly amazed by software that was being built in Stanford Labs 🧪.

There was one particular technology that he was consistently fascinated about: Large Language Models…

This was because of all the applications researchers had built around them. George ended up creating his own LLM product: A very early version of a neural information retrieval model.

He gave it to some of his former students who worked at Morgan Stanley to test out. They used the model to look over this document called the DEFM14A.

The DEFM14A is a 400+ page document that was super important but super gnarly to get through.

People he’d given the software to started copying and pasting the code to give to other friends, and people George had never met before started to use the software.

George had seen that LLMs were working and knew it was the future, so he took a leave of absence from his fully funded PhD program at age 23 to work on Hebbia…

What does Hebbia do? 👨‍💻

Hebbia takes the smartest people in the world who do very silly, mundane and stupid tasks 🙃, and uses LLMs to help them not have to do any of that.

George took inspiration for finding their market niche from Microsoft Excel 👾. Excel took over 90% of the fixed income industry in the span of 18 months 📈, then people started using it for everything else too.

George started doing the same, first focusing on the private equity due diligence process.

LLMs are good at mapping from one ontology to another ontology. With Private Equity, there’s a data room filled with all companies’ files, and an analyst's job is to fill out 400-question trackers on these businesses.

In Private Equity there is no reporting, it’s all just structured data.

The highest paid people in the field just map data from the data room to this set of criteria to decide whether they’ll pour hundreds of millions of dollars into a company.

Hebbia was aggressive in helping automate this, helping PE firms deploy more capital, quicker, in a more accurate fashion.

This has led them to have now worked with some of the largest asset managers in North America & Europe for due diligence, asset pricing and other research 💸.

Stats📊

On 8 July 2024, Hebbia raised a $130M Series B at a $700M valuation led by Andreesen Horowitz, bringing their total raised to $160M 💸💸. Other investors included Index Ventures, Google Ventures, Peter Thiel, and Jerry Yang (founder of Yahoo).

📈 They’ve done $13M in ARR profitably this year, up 15X over the last 18 months.

🤔 Their growth is impressive but is a 50x ARR valuation really justified?

Over the past 18 months they’ve also 5xed their headcount and drove 2% of OpenAI’s daily volume.

The founder believes that Hebbia will be the most important software product for the next 100 years, and says that they’re not settling for anything less.

This has earned George, at the age of 25, a spot in the Forbes 30 Under 30 list for Enterprise Technology.

Before you go 👋 

Prove safe and responsible use of AI - get $1,000 off Vanta

If you’re building or selling AI-powered products, demonstrating top-notch security practices and establishing trust is more important than ever.

With Vanta, you can quickly and easily demonstrate compliance with gold-standard AI frameworks like ISO 42001 and NIST AI RMF.

Vanta helps you prove secure deployment of AI, build customer trust, and accelerate your sales cycle.

Plus, with Vanta’s Questionnaire Automation, you can automate your responses to lengthy security questionnaires about your security posture and AI practices.

Learn more and claim a special offer of $1,000 off Vanta at the link below.

Get $1,000 off Vanta

Thank you for reading.

Remember, if you know anyone who might like this newsletter, sharing is caring.